Some of Carnegie Mellon’s best minds made a cartoon duck move with their fingers.
What if you could take the grunt work out of animation? With the aid of a small motion sensor, a team of Carnegie Mellon researchers have done just that, enabling any still image to be instantly animated with the flick of the wrist.
In a video uploaded by Ali Momeni, a faculty member at Carnegie Mellon University's School of Art Fabrication, Momeni described Dranimate, his team's animation process. A program defines the edges of a hand-drawn image, and the image is converted to a mesh composed of tetrahedral elements. Then, using a Leap motion sensor, a user maps their fingers to the articulated appendages of the drawing. If the drawing was of a humanoid, for example, the user could map their fingers to the arms, legs and head of the drawing.
What results is akin to a digital marionette, where each finger's location and movement above the sensor is converted to the x-axis and y-axis coordinates of an associated extension of the drawing. A user can twiddle their fingers to exact an intuitive level of control over the drawing.
Animation has come a long way in complexity, realism, and ease of access over the past three decades, thanks in no small part to advancements in computing. But the actual process of animating still images—by drawing them over and over with small changes—remains a relatively time consuming process. For certain types of animation and digital storytelling, however, Dranimate looks like an accessible alternative.